1. Field of the Invention
This invention relates generally to medical device systems and, more particularly, to medical device systems and methods capable of estimating the risk of death (an extreme event) at various time-scales and managing said risk in a patient with seizures and epilepsy.
2. Description of the Related Art
Standardized mortality rates in epilepsy patients are higher than in the general population by two to threefold. Relative survivorship (RS) following the diagnosis of epilepsy decreases as the time from diagnosis increases (91% after 5 years; 85% after 10 years; 83% after 15 years. In patients with frequent and severe seizures (pharmaco-resistant epilepsy) mortality is much higher than in those with infrequent seizures, with standardized mortality ratios ranging from 2.2-6.5. Even higher mortality is found in patients with epilepsy and congenital or peri-natal neurological deficits with standardized mortality ratios between 7 and 50.
The various teachings herein conflate new knowledge into embodiments to fill the technology void for prevention of premature death and the crippling neurological sequelae due to late intervention, while addressing the limitations of the existing art which has little if any clinical value. For example, unlike U.S. PUB 2010/0198289, embodiments described herein estimate the risk of death not just of SUPED which although the most dramatic form, accounts for less than ⅕th of all epilepsy related deaths. This application addresses said limitations of the art, using a multivariate (autonomic, neurologic, metabolic signals, etc.) approach at multiple time scales using past behavior (system's history) not just univariately (breathing), at one time scale taking that takes into consideration only one event in the present as U.S. PUB 2010/019828. The embodiments herein incorporate not just rate but also rhythm, pattern and morphology of not just respiratory but also of cardiac and other autonomic activity and neurologic activity which increase predictive power of risk of death; the embodiments herein take into account multiple state factors, the environment and the condition of the patient which may increase risk death, unlike US PUB 2010/0198289 which limits the search to seizures; the embodiments herein use seizure severity, inter-seizures interval, site of seizure origin, etc.; the embodiments herein test arousal and responsiveness which not only increases sensitivity and specificity of risk of death estimates but does not ignore (for performance optimization) that death in epilepsy is not restricted to sleep and in over 80% is not unexplained; the embodiments herein exploit the “built-in” protection against cardiac or respiratory depression/arrest afforded by increased levels of reticulo-thalamo-cortical reticular activation by upwardly modulating them when necessary. At a more fundamental level, the practice of U.S. PUB 2010/0198289 exhibits further deficiencies. 1. Unlike the embodiments herein, it has no power to discriminate within a clinically useful/safe time, apneas and hypopneas occurring in the context of obstructive (or central) sleep apnea syndrome from those of SUDEP. This will result in potentially tens/hundreds of false positive detections each night, rendering that approach useless (patients and caregivers will reject any method that issues multiple nightly false warning of impending death). Given the embodiments of that application (all univariate and without regard for the system's history) the only way to decrease the intolerable (to patients and caregivers) large number of false positives is to increase the SUDEP index threshold to such a high level that the risk will not be identified, if at all, until it will be too late for prevention of death or of hypoxic-ischemic brain damage. Specifically, U.S. PUB 2010/0198289, unlike the instant disclosure, cannot accurately distinguish (until it is too late) apneas of SUDEP or of death from those associated with sleep apnea syndromes; 2. Its treatment of seizures is likely to increase, not decrease the risk of death because: a) “vagal nerve stimulation modifies synchronization between cardiac and respiratory activity, resulting in poor optimization of oxygen delivery to tissues that can be regarded as an additive side effect, which should be considered in patients with already altered brain function. This interaction between the effects of VNS and potential autonomic nervous system dysfunction already reported in epileptic patients should be considered to be potentially life-threatening” (Epilepsia 2009; 50(11):2473-80); b) it increases the resistance of the upper airways to the passage of air (Epilepsy Res. 010; 89(2-3):227-31) and causes bronchial hyperreactivity and increase in secretions, both of which reduce gas exchange (Bull Eur Physiopathol Respir. 1986; 22 Suppl. 7:112-42).
Sleep apnea syndromes which are more prevalent than epilepsy manifest with hypopneas and apneas that recur aperiodically and frequently (tens to hundreds of events each night). Estimating the risk of SUDEP in patients with epilepsy and these syndromes is fraught with uncertainty and thus of little clinical value if this estimation is based only on a short monitoring window (one event at a time) and only one biologic signal (respirations). The embodiments herein overcome these serious limitations by using windows of various lengths (10 s to months) and incorporating into the estimate of risk of death, features such as apnea duration, magnitude, frequency, inter-apnea intervals as a function of sleep cycle, treatment with CNS depressant drugs (that blunt alertness), fitness level and health status of a patient among others. By characterizing (over days to weeks) these variables and building for example probability density functions, it is then and only then, possible to issue useful estimates of the risk of SUDEP.
This is the first invention to utilize not only neurologic, autonomic and other body signals but do so at informative and therapeutically useful time scales, in a multi-variate adaptive manner to optimize sensitivity and specificity of detection of factors correlated with or causative of death in epilepsy and to institute therapies that prevent a fatal outcome as well as irreversible damage to body organs.